Background A large number of studies have shown that manganese can affect the growth and development of human body. The infant stage is a critical period of rapid growth and development, and it is more susceptible to the toxic effects due to manganese exposure.

Objective The study is designed to investigate the relationship between umbilical serum manganese level and the weight of one-year-old infants.

Methods The present study was based on the Laizhou Wan Birth Cohort Study. A total of 183 mother-child pairs were finally enrolled, who completed questionnaires and umbilical cord blood measurement. The participants' demographic characteristics and one-year-old infants' diet were collected, and infants' height, weight, and umbilical cord blood levels of manganese, plead, arsenic, and iron were measured. The study subjects were divided into four groups (Q1-Q4) according to quartiles of manganese exposure level, and the general demographic characteristics were described. Univariate regression analysis was used to study the influencing factors of infants' weight at one year old. After adjusting for pregnant woman's body mass index (BMI), family monthly income, parity, infant's weight at birth, infant's sex, infant's staple food intake frequency, lead, iron, and arsenic in umbilical cord blood, generalized linear model was used to analyze the relationship between umbilical blood manganese level and the weight of one-year-old infants, and the effect on different genders was further studied.

Results The average age of the 183 pregnant women was (27.90±4.61) years old, and the pre-pregnancy BMI was (21.72±3.04) kg·m^-2. Among them, 132 pregnant women were primiparas (72.1%), and 136 pregnant women had no active or passive smoking behaviors (74.3%). Most infants of one year old often consumed staple foods (83.6%). The one-year-old infants weighed (10.55±1.15) kg, and was (76.97±2.71) cm in height. The demographic characteristics of pregnant women showed no significant differences among Q1 to Q4 umbilical cord blood manganese concentration groups (P>0.05), while the weight and height of the one-year-old infants showed significant differences (F=4.111, P=0.008; F=2.717, P=0.046). The positive rates of manganese, lead, iron, and arsenic in umbilical cord blood samples were 82.5% (151/183), 85.8% (157/183), 99.5% (182/183), and 79.8% (146/183), respectively. The median umbilical cord blood concentrations of manganese, lead, iron, and arsenic were 106.1 μg·L^-1, 36.3 μg·L^-1, 462.7 mg·L^-1, and 6.4 μg·L^-1, respectively. The correlations of the weight of one-year-old infants were significant with the concentration of manganese in umbilical cord blood (P=0.005), the concentration of lead in umbilical cord blood (P=0.030), the frequency of staple food intake of one-year-old infants (P < 0.05), and the sex of infants (P < 0.001). For each increase of a logarithmic unit of umbilical blood manganese, the average weight of selected one-year-old infants was reduced by 0.10 kg (95% CI:-0.12--0.07, P=0.021), that of the Q2 group was reduced by 0.68 kg (95% CI:-1.11--0.26, P=0.002), while there was no such an association in the Q3 group or the Q4 group. After sex stratification, for every logarithmic unit increase of umbilical blood manganese, the average weight of the one-year-old boys was reduced by 0.28 kg (95% CI:-0.42--0.14, P < 0.001), and those of the Q2, Q3, and Q4 boy groups were reduced by 0.81, 0.87, and 0.83 kg (95% CI:-1.42--0.21, P=0.009; 95% CI:-1.49--0.25, P=0.006; 95% CI: -1.46--0.19, P=0.011), respectively. This association was not found in the one-year-old girls.

Conclusion With the increase of umbilical cord blood manganese concentration, the weight of one-year-old infants shows a downward trend, and the effect may be more pronounced in boys than in girls.